AI Automation Engineer

Job Overview

Role & responsibilities

Benchmarking AI/LLM models locally using hugging face

Understanding latency, CPU usage, memory and system constraints

Supporting model selection for use-cases

This is a hands-on engineering role, not a research-only position.

Performance & Latency Analysis:

Measure and Document - Inference latency/CPU//Tokens per second

Identify bottlenecks using structured approach

System observability - tools like top,htop,psutil

explain why latency occurs, not just report numbers

Reporting and Recommendations:

Create clear benchmark tables and summaries

Document trade-offs (quality vs speed vs memory)

Recommend the best model for the use case, with reasoning

Technical Must Have: Python, ML/LLMs, Hugging face transformers, Linux basics, CPU & memory concepts.

Good to Have: Model Quantization (int8/fp16)

GPU Basics (CUDA Awareness)

Profiling & performance tuning

Preferred candidate profile:

Hands on model setup, model result analysis, building and deployment with scripting or tools.

Hands on Hugging face localization or meta-llama or phi-2, model fine tuning, dataset.

What this Role is Not:

Not prompt engineering only

Not just running notebooks